Experimental study on the stitching reinforcement of

composite laminates with a circular hole

X.P. Han
a
, L.X. Li
b,
*
, X.P. Zhu
a
, Z.F. Yue
a
a
Department of Engineering Mechanics, Northwestern Polytechnical University, Xian 710072, China
b
MOE Key Laboratory for Strength and Vibration, Department of Engineering Mechanics, Xian Jiaotong University, Xian 710049, China
Received 3 August 2007; received in revised form 27 January 2008; accepted 10 February 2008
Available online 10 March 2008
Abstract
Experimental study is carried out on the stitching reinforcement of composite laminates containing a circular hole. First, the tensile
strength and stiness are measured, and their dependence on stitching parameters such as stitching needle span, row spacing, edge dis-
tance and stitching type are analyzed. Next, the strain distribution and concentration are investigated analytically and experimentally for
dierent stitching parameters, external load and edge location of the hole. It is shown that the results of stitching reinforcement are quite
dierent for composite laminates with a circular hole, which could provide proper stitching parameters for designers.
2008 Elsevier Ltd. All rights reserved.
Keywords: A. Laminates; A. Layered structures; B. Strength; C. Stress concentrations
1. Introduction
Advanced composite panels with holes are widely used in
aerospace structural components. In such structures, stress
concentration around a hole normally causes failures. As a
typical example, the internal damage induced by stress con-
centration can not be found at the early stage of loading and
may propagate as the load increases. On the other hand, the
type and extent of the damage signicantly aect the
strength, service life and the performance of the composite
structures. Hence, the study on the stress concentration
around a hole in composite laminates is of great practical
importance, and has received considerable attention.
Eiblmeier [1] investigated the inuence of cut-out diam-
eter and reinforcement type on the buckling stability of
CFRP panels. For this purpose, several nite element mod-
els were suggested for dierent loadings, cut-out sizes and
reinforcement rings. Practical tests were also conducted
to validate their numerical results.
Mahdi et al. [2,3] focused on the mechanical perfor-
mance of repaired stiened panels under compression
respectively via the experimental study and the nite ele-
ment analysis. A pristine panel was compared with the pan-
els containing a through hole, as well as with the repaired
panels. The modeling of failure, stiness and strains were
studied. It was shown that the signicant decreases in
strength caused by the damage were almost fully recovered
by the repair scheme used.
Gliesche [4,5] developed the Tailored Fiber Placement
Technology (TFP), and applied it to manufacturing a
stress-eld aligned local reinforcement on an open-hole
tensile plate. The TFP technology could relieve the stress
concentration in a notched composite. Using the nite ele-
ment analysis the stresses in the open-hole plate were deter-
mined under unidirectional tensile loading. The strain
distributions were investigated for dierent loading condi-
tions by the optical deformation measurement method.
The reinforcement of polymer matrix composites by
through-the-thickness stitching is a promising technique
for raising the damage tolerance of laminar structures. In
1990s, Mai et al. [610] published series of work to assess
the potential applications of stitched laminates. Later,
0266-3538/$ - see front matter 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compscitech.2008.02.017
*
Corresponding author. Tel.: +86 29 82668751; fax: +86 29 83237910.
E-mail address: luxianli@mail.xjtu.edu.cn (L.X. Li).
www.elsevier.com/locate/compscitech
Available online at www.sciencedirect.com
Composites Science and Technology 68 (2008) 16491653
COMPOSITES
SCIENCE AND
TECHNOLOGY
Mouritz and Cox [11] proposed an approach to evaluate
the properties of the stitched laminates. These studies
showed that the load-carrying capacity of the laminates
can be improved through a proper stitching reinforcement.
The objective of this paper is to study the stitching rein-
forcement in composite laminates with a circular hole. The
eects of the stitching parameters such as stitching needle
span (the stitch-to-stitch spacing along the hoop direction
around a hole), row spacing (the row-to-row spacing along
the radial direction), edge distance (the minimum radial
distance between the hole edge and the stitch line) are taken
into account. The strains are measured for dierent stitch-
ing parameters, external load and edge location of the hole,
based on which the corresponding strain concentration fac-
tor and the failure strength for this kind of composite lam-
inates are studied.
2. Basic ideas
Assume that r
1
is the stress applied at innity. Thus,
the stress distributions at the hole edge have a maximum
value r
max
. For the purpose of description, the stress con-
centration factor (SCF), say K
r
, is used as usual and
dened as
K
r

r
max
r
1
1
The composite laminate is composed of a number of
layers with dierent material properties, so the stress distri-
butions around the hole are dierent for each layer. Con-
sidering the fact that all of the adjacent layers are well
bonded together, the interlaminar stresses will be ignored
and the strains of the individual layers are assumed to vary
with the laminate. As the common practice, the SCF
around holes in multilayered plates can be expressed as
the weighted average of the SCFs for all layers, that is
K
r

P
m
i1
N
i
K
i
N
2
where N is the gross number of the layers, m is the number
of plies with dierent ber lay-up directions, N
i
is the re-
peated layer number of the ith ply, and K
i
is the SCF in
the ith ply.
The SCF for each layer with a central hole was given by
[12] as
K
1
i
1

E
y
E
x
r
v
yx

E
y
2G
yx

s
3
K
1
i
K
i

31 d=W
2 1 d=W
3

1
2
d
W
M

6
K
1
i
3
1
d
W
M

2
" #
4
where K
i
and K
1
i
denote the SCFs of the nite plate and
the corresponding innite plate, respectively, at the hole
edge whose normal is 90 away from the applied load direc-
tion. E
x
and E
y
are Youngs moduli respectively in x and y
direction, and G
yx
and m
yx
are respectively the shear mod-
ulus and Poissons ratio in xy plane. M, the magnication
factor, is a function of d/w (ratio of hole diameter to plate
width) and can be expressed by
M
2

1 8
31d=W
21d=W
3
1
q
1
2d=W
2
5
It is obvious that K
r
(SCF) will be determined from Eq.
(2) if K
i
is known. However, the solution above mentioned
for the stress concentration around a hole in the multilay-
ered plate is not exact at the hole boundary. In the free-
edge region of the hole, shear stresses exist between various
layers of a laminate. In addition, use of the correction fac-
tor for the niteness of plate width may cause some error.
So, there is room for Eq. (2) still to be improved.The same
analysis can be performed for the strain instead around a
hole. From the maximum strain value e
max
at the hole edge
and the strain e
1
at innity, the strain concentration factor
K
e
can be dened as
K
e

e
max
e
1
6
The strains can be directly observed in specimens. The-
oretically, the deformation in the local region of the hole is
governed by the local stresses at the hole edge, so the stress
concentration factor K
r
and strain concentration factor K
e
are correlated. That is to say, once the strain concentration
at the hole edge is measured, the level of the stress concen-
tration at the hole edge will be determined. Thus, this prob-
lem can be more exactly resolved by eciently measuring
K
e
around the hole with experimental devices.
3. Experimental consideration
In order to study the eect of stitching parameters such
as needle span, row space, edge distance, stitching type (i.e.
single stitching or double stitching), the specimens are
divided into ve types, i.e. three types of single stitching
reinforcement marked respectively by Single-1, Single-2
and Single-3, and two types of double stitching reinforce-
ment marked respectively by Double-1 and Double-2. In
the specimens, the stitching yarns are made of the
Kevalr-49 (1600 denier) and stitched near the hole by use
of the improved lockstitch [13]. In Table 1 are shown
details of the stitching parameters.
The laminate with a circular hole is made of T300/
Epoxy. The diameter of the circular hole is 60 mm. The
ber lay-up in the laminate is [45/45/0/90/0 /0/90/45/
45]
2S
. The reinforced slices, which are made of LY12CZ,
one kind of aluminium alloy, are separately bonded to
the two ends of every specimen. Strain gauges (2540) were
bonded to various locations along the hole edge crosswise,
and connected with a multipass strainometer. In Fig. 1 are
shown the geometry and dimension of specimens.
The two ends of specimens were separately linked up
with the tailored xture and installed in the jaw of the
1650 X.P. Han et al. / Composites Science and Technology 68 (2008) 16491653
testing machine. The design and dimension of the xture are
shown in Fig. 2. The tensile experiments are carried out
using a servo hydraulic material test system (MTS 880-10).
4. Results and discussion
The failure strength and strain of the specimens are
dened as the maximum value of the longitudinal stress
and strain at 0 point around the hole on the stressstrain
curve. Tables 2 and 3 show respectively the failure strength
of specimens and the failure strain of the hole edge for dif-
ferent stitching parameters.
A regular tensile stress-eld is seen at innity far away
from the hole edge and from the location where the
clamped boundary condition is prescribed. The curve of
stress versus strain at innity is plotted in Fig. 3, whereas
the curve of stress versus strain at the hole edge where
the cross-section is minimum is depicted in Fig. 4.
It is seen from Figs. 3 and 4 and Table 2 that the stiness
and the failure strength of the composite laminate with a
hole increases because of stitching reinforcement at the
Table 1
Stitching parameters of stitching reinforcement on the hole
Stitching parameters Single-1 Single-2 Single-3 Double-1 Double-2
Needle span/mm 3 3 5 3 3
Row spacing/mm 0 0 0 5 3
Edge distance/mm 3 5 3 3 3
60
66
76
Stitching lines
300
30
30
200
Fig. 1. Geometry and dimension of specimens.
260
60
25
150
75
20
76
100
80
Fig. 2. Design and dimension of the xture.
Table 2
Failure strength of specimens/MPa
Number Unstitching Single-1 Single-2 Single-3 Double-1 Double-2
1 204.76 201.19 240.24 232.86 241.67 238.10
2 223.81 251.19 239.52 214.29 258.33 226.19
3 182.86 223.10 220.24 220.24 233.33 203.57
4 195.24 240.05 204.29 214.29
5 209.52 209.45 226.19 216.19
Average value 203.81 216.05 229.90 222.46 232.76 219.67
Table 3
Failure strain of hole edge/le
Number Unstitching Single-1 Single-2 Single-3 Double-1 Double-2
1 8204 8629 7230 8195 9027 8139
2 8474 8328 8059 8098 7295 10310
3 7143 8539 8506 7884 9706 8605
4 8619 9374 8874 10405
5 10368 8634 8519 8499
Average value 7940 8896 8361 8059 8684 9192
X.P. Han et al. / Composites Science and Technology 68 (2008) 16491653 1651
hole edge. When reinforced by stitching at the hole edge,
the ber volume fraction is added in thickness, and the
loads can be eectively transferred between the plies in
the local reinforced region. This mechanism improves the
load capacity.
Besides, attention must be paid to the re-distribution of
the stresses in the laminate due to the stitching reinforce-
ment at the hole edge, especially to the stress concentration
and strain concentration newly generated by the closing of
the stitching lines. From Table 3, it is seen that, reinforced
by stitching, the maximum strain at the hole edge becomes
larger, and hence the strain concentration at the hole edge
is increased rather than decreased in this circumstance.
This implies that the so-called notch strengthening may
risk a sudden failure.
Fig. 5 shows the strain concentration factor K
e
with the
distance from the hole edge (along the x axis) at the mini-
mum cross-section around the hole. It is seen from Fig. 5
that the strain concentration at the hole edge is consider-
ably severe either for stitching or for unstitching specimens.
The strain concentration declines gradually with the dis-
tance away from the hole edge, and almost disappears at
the distance of a radius of the hole. For the strain declina-
tion, the scope is roughly the same for both stitching and
unstitching, but the level is dierent.
Fig. 6 depicts the variation of K
e
with load respectively
at 0 point and at 90 point around the hole. From
Fig. 6(b), we see that K
e
at 90 point are strongly dependent
on the load. However, see Fig. 6(a), it is interesting that the
0 1000 2000 3000
0
50
100
150
200
250
S
t
r
e
s
s

/

M
P
a
unstitching
single1
single2
single3
double1
double2
Strain /
Fig. 3. The stressstrain curve (far region).
0 2000 4000 6000 8000 10000 12000
0
50
100
150
200
250
S
t
r
e
s
s

/

M
P
a
unstitching
single1
single2
single3
double1
double2
Strain /
Fig. 4. The stressstrain curve (0at the hole edge).
0 10 20 30 40 50 60
0
1
2
3
4
5
6
7
K
e
unstitching
single1
single2
single3
double1
double2
Distance / mm
Fig. 5. Distribution of K
e
at the minimum cross-section around the hole.
0 2 4 6 8 10 12 14
20
40
60
80
100
120
L
o
a
d
/
K
N
unstitching
single1
single2
single3
double1
double2
(a) 0 point
0 20 40 60 80
0
20
40
60
80
100
120
L
o
a
d
/
K
N
single1
single3
double1
double2
(b) 90 point
K

Fig. 6. Strain concentration factor K

e
at the hole edge.
1652 X.P. Han et al. / Composites Science and Technology 68 (2008) 16491653
strain concentration factor K
e
at 0 point does not vary
with the load for the given specimen (i.e. xing the stitching
parameters). K
e
for unstitching is about 3.3 and increases
after stitching. For the case of the double stitching rein-
forcement, the maximum strain concentration is increased
by 88% around as compared with that of the single stitch-
ing reinforcement. Considering the fact that the ber vol-
ume fraction of the double stitching at the local regions
is much higher than that of the single stitching, the local
strengthening eect becomes more prominent after the
double stitching reinforcement.
By comparing Fig. 6a and b, we also see that K
e
at 90
point is much larger than at 0 point. The reasons are two-
folded. On one hand, the vertical and horizontal load
aects each other due to the eect of the Poissons ratio
at the local regions, where the material bore shrinks freely.
This mechanism causes the compression deformation to
quite a large extent. On the other hand, with the material
shrinking freely, the vertical and horizontal deformations
are no longer linearly dependent on the load after a critical
value. According to the denition, K
e
does not remain a
constant under this circumstance, but changes with the
applied load and the local deformation.
5. Conclusions
From the current work, we conclude that the stiness
and the failure strength of the composite laminate are
increased after stitching reinforcement. There exists a notch
strengthening eect, which implies that the maximum
strain at the hole edge becomes higher. Comparing the
stitching reinforcement with unstitching, the scope of strain
declination is roughly the same, but the declination level is
obviously increased. The strain concentration factor K
e
at
90 point around the hole varies with load, whereas K
e
at
0 point keep the relevant constant for a given specimen.
For anisotropic laminates K
e
is various at dierent point
around the hole, unlike the isotropic plates. Combining
the results in all respects, the Single-2 and the Double-1
are the better choices for the composite laminates when
reinforced by stitching on the hole.
Acknowledgements
This work was supported by the National Natural Sci-
ence Foundation of China (10472090, 10572109), the
New Century Excellent Talents of Ministry of Education
(NCET-04-0930) and China 973 Program (2007CB707705).
References
[1] Eiblmeier J, Loughlan J. The buckling response of carbon ber
composite panels with reinforced cut-outs. Compos Struct 1995;32:
97113.
[2] Mahdi S, Kinloch AJ, Matthews FL. The mechanical performance of
repaired stiened panels, Part I: experimental characterization.
Compos Part B 2002;33(5):34354.
[3] Mahdi S, Kinloch AJ, Criseld MA. The mechanical performance of
repaired stiened panels, Part II: nite element modeling. Compos
Part B 2002;33(5):35566.
[4] Gliesche K, Feltin D. Automated manufacture of textile reinforce-
ments by tailored ber placement. In: International symposium on
composites manufacturing for aircraft structures, Flevoland 2002; p.
3031.
[5] Gliesche K, Hubner T, Orawetz H. Application of the tailored ber
placement process for a local reinforcement on an open-hole tensile
plate from carbon/epoxy laminates. Compos Sci Technol 2003;63(1):
818.
[6] Jain LK, Mai YW. Analysis of stitched laminated ENF specimens for
interlaminar mode fracture toughness. Int J Fracture 1994;68:21944.
[7] Jain L K, Mai Y-W. On the eect of stitching on mode I delamination
toughness of laminated composites. Compos Sci Technol 1994;51:
33145.
[8] Jain L K, Mai Y-W. Determination of mode II delamination
toughness of stitched laminated composites. Compos Sci Technol
1995;55:24153.
[9] Dranseld KA, Jain LK, Mai Y-W. On the eects of stitching in
CFRPs-I. Mode I delamination toughness. Compos Sci Technol
1998;58:81527.
[10] Jain LK, Dranseld KA, Mai Y-W. On the eects of stitching in
CFRPs-I. Mode II delamination toughness. Compos Sci Technol
1998;58:82937.
[11] Mouritz A P, Cox B N. A mechanistic approach to the properties of
stitched laminates. Compos Part A 2000;31:127.
[12] Wu HC, Mu B. On stress concentrations for isotropic/orthotropic
plates and cylinders with a circular hole. Compos Part B 2003;34:
12734.
[13] Sankey BV, Zhu H. The eect of stitching on the low-velocity impact
response of delaminated composite beams. Compos Sci Technol
2000;60:268191.
X.P. Han et al. / Composites Science and Technology 68 (2008) 16491653 1653